Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly)
Python
minor: Add Atlas Cloud provider (#2035)
nb213b920b8db7939
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.mdindex 01dde124..3ebaf366 100644--- a/CONTRIBUTING.md+++ b/CONTRIBUTING.md@@ -4,7 +4,8 @@ Thanks for helping improve Free Claude Code. Keep changes focused, test the beha## Before Opening A Pull Request-- Open an issue before proposing README changes.+- Standalone documentation PRs are reserved for the repository owner.+- Include relevant README and other documentation updates with production code changes. These accompanying updates do not require a separate issue.- Do not open Docker integration pull requests.- For bugs, include every model mapping, the active model when the failure occurred, the complete error, and reproducible steps.- Add focused tests for behavior changes and relevant edge cases.diff --git a/README.md b/README.mdindex 7e1cdd62..e4eff26b 100644--- a/README.md+++ b/README.md@@ -25,7 +25,7 @@## What You Get-- **59 ToS-friendly providers. 1.3B+ free tokens every month.** Use free, paid, subscription, and local models from one searchable UI without putting your account at risk. FCC follows provider terms and removes integrations if they stop being allowed.+- **61 ToS-friendly providers. 1.3B+ free tokens every month.** Use free, paid, subscription, and local models from one searchable UI without putting your account at risk. FCC follows provider terms and removes integrations if they stop being allowed.- **11 coding agents. One model catalog.** Run [Claude Code](https://code.claude.com/docs/en/overview), [Codex](https://github.com/openai/codex), [Pi](https://github.com/earendil-works/pi), [OpenCode](https://github.com/anomalyco/opencode), [Cline](https://github.com/cline/cline), [Hermes](https://github.com/NousResearch/hermes-agent), [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness), [Grok Build](https://github.com/xai-org/grok-build), [Muse Code](https://research.meta.ai/blog/introducing-muse-code-and-muse-spark-1-2/), [Aider](https://aider.chat/), or [VS Code Chat](https://code.visualstudio.com/) with your FCC models.- **Keep coding through provider outages.** After retries are exhausted, FCC automatically tries your next configured model without making you restart the turn. This works across every client.- **Up to 90% fewer terminal-output tokens.** Optional [RTK](https://github.com/rtk-ai/rtk) filters common command output, while five FCC optimizations handle quota probes, command-prefix detection, titles, suggestions, and filepaths without calling a provider.@@ -269,6 +269,7 @@ For another API endpoint, open **Providers → Custom providers → Add provider| [Experiential Labs](https://platform.experientiallabs.ai/) | `EXPLABS_API_KEY` | `experiential/union-alpha` || [Cheaper Inference](https://cheaperinference.com/signup) | `CHEAPER_INFERENCE_API_KEY` | `cheaperinference/gpt-5.4-mini` || [OrcaRouter](https://www.orcarouter.ai/) | `ORCAROUTER_API_KEY` | `orcarouter/deepseek/deepseek-v4-flash-free` |+| [Atlas Cloud](https://console.atlascloud.ai/api-keys) | `ATLASCLOUD_API_KEY` | `atlascloud/deepseek-ai/deepseek-v4-flash` || [xKiro](https://xkiro.com/) | `XKIRO_API_KEY` | `xkiro/qwen/qwen3.7-flash:free` || [Opper](https://platform.opper.ai/) | `OPPER_API_KEY` | `opper/claude-sonnet-4-6` || [Ollama Cloud](https://ollama.com/settings/keys) | `OLLAMA_API_KEY` | `ollama_cloud/qwen3-coder:480b` |diff --git a/smoke/lib/config.py b/smoke/lib/config.pyindex 116c51eb..e62d9cf2 100644--- a/smoke/lib/config.py+++ b/smoke/lib/config.py@@ -99,6 +99,7 @@ PROVIDER_SMOKE_DEFAULT_MODELS: dict[str, str] = {"experiential": "experiential/union-alpha","cheaperinference": "cheaperinference/gpt-5.4-mini","orcarouter": "orcarouter/deepseek/deepseek-v4-flash-free",+ "atlascloud": "atlascloud/deepseek-ai/deepseek-v4-flash","xkiro": "xkiro/qwen/qwen3.7-flash:free","agnes": "agnes/agnes-2.0-flash","zenmux": "zenmux/deepseek/deepseek-v4-flash-free",diff --git a/src/free_claude_code/api/admin_static/providers/SOURCES.md b/src/free_claude_code/api/admin_static/providers/SOURCES.mdindex 8cb7d53b..d47d0a74 100644--- a/src/free_claude_code/api/admin_static/providers/SOURCES.md+++ b/src/free_claude_code/api/admin_static/providers/SOURCES.md@@ -43,6 +43,7 @@ The LLaMA.cpp icon comes from the upstream llama.cpp repository; its MIT license| `opencode.svg` | https://unpkg.com/@lobehub/[email protected]/icons/opencode.svg || `openrouter.svg` | https://unpkg.com/@lobehub/[email protected]/icons/openrouter.svg || `orcarouter.png` | https://www.orcarouter.ai/apple-touch-icon.png |+| `atlascloud.svg` | https://unpkg.com/@lobehub/[email protected]/icons/atlascloud.svg || `xkiro.png` | https://xkiro.com/images/logo/logo-xt-green.png || `opper.svg` | https://opper.ai/ (provided by Opper, single-colour `currentColor` mark) || `poolside-color.svg` | https://unpkg.com/@lobehub/[email protected]/icons/poolside-color.svg |diff --git a/src/free_claude_code/api/admin_static/providers/atlascloud.svg b/src/free_claude_code/api/admin_static/providers/atlascloud.svgnew file mode 100644index 00000000..f8479194--- /dev/null+++ b/src/free_claude_code/api/admin_static/providers/atlascloud.svg@@ -0,0 +1 @@+<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Atlas Cloud</title><path d="M20.2,18.01L12,0.47,3.8,18.01l-2.58,5.52c1.62,-1.05,3.39,-1.86,5.26,-2.41,1.76,-0.51,3.61,-0.79,5.52,-0.79,0.98,0,1.95,0.08,2.9,0.22l-1.86,-4.3c-0.53,-0.1,-2.87,-0.1,-4.59,0.3l3.56,-8.28,5.52,12.85c0.01,0,0.02,0.01,0.03,0.01,1.86,0.55,3.62,1.36,5.23,2.4l-2.58,-5.52Z"></path></svg>\ No newline at end of filediff --git a/src/free_claude_code/config/admin/provider_manifest.py b/src/free_claude_code/config/admin/provider_manifest.pyindex 341de97c..54ac1f5e 100644--- a/src/free_claude_code/config/admin/provider_manifest.py+++ b/src/free_claude_code/config/admin/provider_manifest.py@@ -300,6 +300,13 @@ _PROVIDER_FIELD_OVERRIDES: dict[str, ProviderFieldOverride] = {"www.orcarouter.ai/console."),},+ "ATLASCLOUD_API_KEY": {+ "label": "Atlas Cloud API Key",+ "description": (+ "Atlas Cloud OpenAI-compatible multi-provider gateway API key for "+ "api.atlascloud.ai/v1; create one at console.atlascloud.ai/api-keys."+ ),+ },"XKIRO_API_KEY": {"label": "xKiro API Key","description": (diff --git a/src/free_claude_code/config/provider_catalog.py b/src/free_claude_code/config/provider_catalog.pyindex 85e6da80..2ca27432 100644--- a/src/free_claude_code/config/provider_catalog.py+++ b/src/free_claude_code/config/provider_catalog.py@@ -91,6 +91,8 @@ EXPERIENTIAL_DEFAULT_BASE = "https://api.experientiallabs.ai/v1"CHEAPERINFERENCE_DEFAULT_BASE = "https://api.cheaperinference.com/v1"# OrcaRouter OpenAI-compatible multi-provider gateway.ORCAROUTER_DEFAULT_BASE = "https://api.orcarouter.ai/v1"+# Atlas Cloud OpenAI-compatible multi-provider gateway.+ATLASCLOUD_DEFAULT_BASE = "https://api.atlascloud.ai/v1"# xKiro OpenAI-compatible multi-provider gateway.XKIRO_DEFAULT_BASE = "https://api.xkiro.com/v1"# Opper OpenAI-compatible Chat Completions gateway.@@ -743,6 +745,17 @@ PROVIDER_CATALOG: dict[str, ProviderDescriptor] = {default_base_url=ORCAROUTER_DEFAULT_BASE,proxy_attr="orcarouter_proxy",),+ "atlascloud": ProviderDescriptor(+ provider_id="atlascloud",+ display_name="Atlas Cloud",+ website_url="https://www.atlascloud.ai/",+ logo_filename="atlascloud.svg",+ credential_env="ATLASCLOUD_API_KEY",+ credential_url="https://console.atlascloud.ai/api-keys",+ credential_attr="atlascloud_api_key",+ default_base_url=ATLASCLOUD_DEFAULT_BASE,+ proxy_attr="atlascloud_proxy",+ ),"xkiro": ProviderDescriptor(provider_id="xkiro",display_name="xKiro",diff --git a/src/free_claude_code/config/settings.py b/src/free_claude_code/config/settings.pyindex f2c1a68d..65d34889 100644--- a/src/free_claude_code/config/settings.py+++ b/src/free_claude_code/config/settings.py@@ -244,6 +244,11 @@ class Settings(BaseModel):default=None, validation_alias="ORCAROUTER_API_KEY")+ # ==================== Atlas Cloud (OpenAI-compatible gateway) ====================+ atlascloud_api_key: OptionalNonEmptyString = Field(+ default=None, validation_alias="ATLASCLOUD_API_KEY"+ )+# ==================== xKiro (OpenAI-compatible gateway) ====================xkiro_api_key: OptionalNonEmptyString = Field(default=None, validation_alias="XKIRO_API_KEY"@@ -583,6 +588,9 @@ class Settings(BaseModel):orcarouter_proxy: OptionalNonEmptyString = Field(default=None, validation_alias="ORCAROUTER_PROXY")+ atlascloud_proxy: OptionalNonEmptyString = Field(+ default=None, validation_alias="ATLASCLOUD_PROXY"+ )xkiro_proxy: OptionalNonEmptyString = Field(default=None, validation_alias="XKIRO_PROXY")diff --git a/src/free_claude_code/providers/openai_chat/profiles.py b/src/free_claude_code/providers/openai_chat/profiles.pyindex f8015a3e..c3f445c1 100644--- a/src/free_claude_code/providers/openai_chat/profiles.py+++ b/src/free_claude_code/providers/openai_chat/profiles.py@@ -754,6 +754,37 @@ OPENAI_CHAT_PROFILES: dict[str, OpenAIChatProfile] = {optional_sequence_items=(("supported_endpoint_types", "openai"),),),),+ # Measured against api.atlascloud.ai/v1 across DeepSeek, GLM, Kimi, Qwen, MiniMax,+ # OpenAI, Anthropic, Google and xAI models: only ``high`` and ``xhigh`` are+ # accepted by every model (``none`` is a 400 for glm-5.3, minimax-m2.7 and+ # gpt-5.5, ``low``/``medium`` for qwen3.5-plus, ``max`` for several), so the+ # efforts map onto those two and "off" sends no field.+ "atlascloud": OpenAIChatProfile(+ _policy(+ "ATLASCLOUD",+ ReasoningReplayMode.REASONING_CONTENT,+ default_max_tokens=ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS,+ ),+ NamedEffortReasoning(+ (+ (ReasoningEffort.MINIMAL, "high"),+ (ReasoningEffort.LOW, "high"),+ (ReasoningEffort.MEDIUM, "high"),+ (ReasoningEffort.HIGH, "high"),+ (ReasoningEffort.XHIGH, "xhigh"),+ (ReasoningEffort.MAX, "xhigh"),+ ),+ enabled_value="high",+ ),+ model_listing=OpenAIModelListing(+ path="/models",+ # ``supported_features`` under-reports reasoning (deepseek-v4-flash lists+ # only tools yet returns reasoning tokens), so thinking is left unknown.+ input_modalities_path=("input_modalities",),+ context_window_tokens_path=("context_length",),+ max_output_tokens_path=("max_output_length",),+ ),+ ),"xkiro": OpenAIChatProfile(_policy("XKIRO",diff --git a/tests/contracts/test_provider_catalog_order.py b/tests/contracts/test_provider_catalog_order.pyindex fb340790..ff181216 100644--- a/tests/contracts/test_provider_catalog_order.py+++ b/tests/contracts/test_provider_catalog_order.py@@ -60,6 +60,7 @@ _EXPECTED_PROVIDER_ORDER: tuple[str, ...] = ("experiential","cheaperinference","orcarouter",+ "atlascloud","xkiro","opper","ollama_cloud",diff --git a/tests/providers/test_atlascloud.py b/tests/providers/test_atlascloud.pynew file mode 100644index 00000000..724a13ee--- /dev/null+++ b/tests/providers/test_atlascloud.py@@ -0,0 +1,265 @@+"""Tests for the Atlas Cloud OpenAI-chat gateway profile and catalog."""++from types import SimpleNamespace+from unittest.mock import AsyncMock++import httpx2+import pytest+from openai import AsyncOpenAI++from free_claude_code.application.model_metadata import ProviderModelInfo+from free_claude_code.config.constants import ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS+from free_claude_code.config.provider_catalog import ATLASCLOUD_DEFAULT_BASE+from free_claude_code.core.anthropic.models import MessagesRequest+from free_claude_code.core.json_types import JsonObject, JsonValue+from free_claude_code.core.model_capabilities import ModelInputModality+from free_claude_code.core.reasoning import ReasoningEffort, ReasoningPolicy+from free_claude_code.providers.model_listing import ModelListResponseError+from free_claude_code.providers.openai_chat import OpenAIChatProvider+from tests.providers.support import (+ REASONING_DEFAULT,+ REASONING_OFF,+ REASONING_ON,+ immediate_admission,+ make_provider_config,+ profiled_provider,+ reasoning_for,+)++_MODEL = "deepseek-ai/deepseek-v4-flash"+++def atlascloud_provider() -> OpenAIChatProvider:+ return profiled_provider(+ "atlascloud",+ make_provider_config(+ api_key="test-atlascloud-key",+ base_url=ATLASCLOUD_DEFAULT_BASE,+ ),+ admission=immediate_admission(provider_name="atlascloud", max_attempts=1),+ )+++def _request(**overrides: JsonValue) -> MessagesRequest:+ payload: JsonObject = {+ "model": _MODEL,+ "messages": [{"role": "user", "content": "Inspect the file."}],+ "tools": [+ {+ "name": "read_file",+ "description": "Read a file",+ "input_schema": {+ "type": "object",+ "properties": {"path": {"type": "string"}},+ "required": ["path"],+ },+ }+ ],+ }+ payload.update(overrides)+ return MessagesRequest.model_validate(payload)+++def test_constructs_standard_openai_chat_provider(+ atlascloud_provider: OpenAIChatProvider,+) -> None:+ assert isinstance(atlascloud_provider, OpenAIChatProvider)+ assert atlascloud_provider._provider_name == "ATLASCLOUD"+ assert atlascloud_provider._api_key == "test-atlascloud-key"+ assert atlascloud_provider._base_url == ATLASCLOUD_DEFAULT_BASE+++ ("reasoning", "expected"),+ [+ (REASONING_DEFAULT, None),+ (REASONING_OFF, None),+ (REASONING_ON, "high"),+ (ReasoningPolicy.on(effort=ReasoningEffort.MINIMAL), "high"),+ (ReasoningPolicy.on(effort=ReasoningEffort.LOW), "high"),+ (ReasoningPolicy.on(effort=ReasoningEffort.MEDIUM), "high"),+ (ReasoningPolicy.on(effort=ReasoningEffort.HIGH), "high"),+ (ReasoningPolicy.on(effort=ReasoningEffort.XHIGH), "xhigh"),+ (ReasoningPolicy.on(effort=ReasoningEffort.MAX), "xhigh"),+ ],+)+def test_encodes_only_efforts_every_model_accepts(+ atlascloud_provider: OpenAIChatProvider,+ reasoning: ReasoningPolicy,+ expected: str | None,+) -> None:+ body = atlascloud_provider._chat._build_request_body(+ _request(),+ reasoning=reasoning,+ )++ assert body.get("reasoning_effort") == expected+ assert body["max_tokens"] == ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS+ assert body["model"] == _MODEL+ assert body["tools"][0]["function"]["name"] == "read_file"+ assert "thinking" not in body+ assert "extra_body" not in body+++def test_replays_reasoning_content_with_tool_history(+ atlascloud_provider: OpenAIChatProvider,+) -> None:+ request = _request(+ messages=[+ {"role": "user", "content": "Inspect the file."},+ {+ "role": "assistant",+ "content": [+ {"type": "thinking", "thinking": "Read it first."},+ {"type": "text", "text": "I will inspect it."},+ {+ "type": "tool_use",+ "id": "toolu_1",+ "name": "read_file",+ "input": {"path": "example.py"},+ },+ ],+ },+ {+ "role": "user",+ "content": [+ {+ "type": "tool_result",+ "tool_use_id": "toolu_1",+ "content": "print('hello')",+ }+ ],+ },+ ]+ )++ body = atlascloud_provider._chat._build_request_body(+ request,+ reasoning=reasoning_for(request),+ )++ assert body["messages"][1] == {+ "role": "assistant",+ "content": "I will inspect it.",+ "reasoning_content": "Read it first.",+ "tool_calls": [+ {+ "id": "toolu_1",+ "type": "function",+ "function": {+ "name": "read_file",+ "arguments": '{"path": "example.py"}',+ },+ }+ ],+ }+ assert body["messages"][2] == {+ "role": "tool",+ "tool_call_id": "toolu_1",+ "content": "print('hello')",+ }+ assert (+ atlascloud_provider._profile.reasoning_delta(+ SimpleNamespace(reasoning_content="next thought")+ )+ == "next thought"+ )+++async def test_model_catalog_maps_context_output_and_input_modalities(+ atlascloud_provider: OpenAIChatProvider,+) -> None:+ atlascloud_provider._client.get = AsyncMock(+ return_value={+ "data": [+ {+ "id": _MODEL,+ "context_length": 1048576,+ "max_output_length": 65536,+ "input_modalities": ["text"],+ # Atlas Cloud lists no "reasoning" here although the model+ # returns reasoning tokens, so thinking must stay unknown.+ "supported_features": ["json_mode", "tools"],+ },+ {+ "id": "moonshotai/kimi-k3",+ "context_length": 1048576,+ "input_modalities": ["text", "image", "video"],+ },+ ]+ }+ )++ model_infos = await atlascloud_provider.list_model_infos()++ assert model_infos == frozenset(+ {+ ProviderModelInfo(+ _MODEL,+ input_modalities=frozenset({ModelInputModality.TEXT}),+ context_window_tokens=1048576,+ max_output_tokens=65536,+ ),+ ProviderModelInfo(+ "moonshotai/kimi-k3",+ input_modalities=frozenset(+ {ModelInputModality.TEXT, ModelInputModality.IMAGE}+ ),+ context_window_tokens=1048576,+ ),+ }+ )+ atlascloud_provider._client.get.assert_awaited_once_with(+ "/models",+ cast_to=object,+ )+++ ("payload", "message"),+ [+ ({"models": []}, "expected top-level data array"),+ ({"data": [{}]}, "include id"),+ ({"data": []}, "did not include any model ids"),+ ],+)+async def test_rejects_malformed_or_empty_catalog_atomically(+ atlascloud_provider: OpenAIChatProvider,+ payload: object,+ message: str,+) -> None:+ atlascloud_provider._client.get = AsyncMock(return_value=payload)++ with pytest.raises(ModelListResponseError, match=message):+ await atlascloud_provider.list_model_infos()+++async def test_model_catalog_uses_documented_url_and_bearer_auth(+ atlascloud_provider: OpenAIChatProvider,+) -> None:+ requests: list[httpx2.Request] = []++ def handler(request: httpx2.Request) -> httpx2.Response:+ requests.append(request)+ return httpx2.Response(200, json={"data": [{"id": _MODEL}]})++ await atlascloud_provider._client.close()+ atlascloud_provider._client = AsyncOpenAI(+ api_key="wire-atlascloud-key",+ base_url=ATLASCLOUD_DEFAULT_BASE,+ max_retries=0,+ http_client=httpx2.AsyncClient(transport=httpx2.MockTransport(handler)),+ )+ try:+ model_infos = await atlascloud_provider.list_model_infos()+ finally:+ await atlascloud_provider.cleanup()++ assert model_infos == frozenset({ProviderModelInfo(_MODEL)})+ assert len(requests) == 1+ assert str(requests[0].url) == "https://api.atlascloud.ai/v1/models"+ assert requests[0].headers["authorization"] == "Bearer wire-atlascloud-key"diff --git a/tests/providers/test_provider_runtime.py b/tests/providers/test_provider_runtime.pyindex e38de52c..9a5b43ff 100644--- a/tests/providers/test_provider_runtime.py+++ b/tests/providers/test_provider_runtime.py@@ -13,6 +13,7 @@ from free_claude_code.application.errors import (from free_claude_code.config.nim import NimSettingsfrom free_claude_code.config.provider_catalog import (AGNES_DEFAULT_BASE,+ ATLASCLOUD_DEFAULT_BASE,BEDROCK_DEFAULT_BASE,CHEAPERINFERENCE_DEFAULT_BASE,CHUTES_DEFAULT_BASE,@@ -131,6 +132,7 @@ def _make_settings(**overrides):mock.experiential_api_key = "test_experiential_key"mock.cheaperinference_api_key = "test_cheaperinference_key"mock.orcarouter_api_key = "test_orcarouter_key"+ mock.atlascloud_api_key = "test_atlascloud_key"mock.xkiro_api_key = "test_xkiro_key"mock.nvidia_nim_proxy = Nonemock.open_router_proxy = None@@ -182,6 +184,7 @@ def _make_settings(**overrides):mock.experiential_proxy = Nonemock.cheaperinference_proxy = Nonemock.orcarouter_proxy = None+ mock.atlascloud_proxy = Nonemock.xkiro_proxy = Nonemock.kilo_api_key = "test_kilo_key"mock.kilo_proxy = None@@ -416,6 +419,36 @@ async def test_orcarouter_provider_config_uses_key_base_and_proxy() -> None:assert isinstance(provider, OpenAIChatProvider)+async def test_atlascloud_provider_config_uses_key_base_and_proxy() -> None:+ descriptor = PROVIDER_CATALOG["atlascloud"]+ settings = _make_settings(+ atlascloud_api_key="atlascloud-token",+ atlascloud_base_url="https://unused.example/v1",+ atlascloud_proxy="http://proxy.test:8080",+ )++ config = build_provider_config(descriptor, settings)+ with patch("free_claude_code.providers.openai_chat.client.AsyncOpenAI"):+ provider = await create_provider(+ "atlascloud",+ settings,+ ProviderAdmissionRegistry(ProviderAdmissionLimits.from_settings(settings)),+ )++ assert descriptor.display_name == "Atlas Cloud"+ assert descriptor.credential_env == "ATLASCLOUD_API_KEY"+ assert descriptor.credential_attr == "atlascloud_api_key"+ assert descriptor.credential_url == "https://console.atlascloud.ai/api-keys"+ assert descriptor.default_base_url == ATLASCLOUD_DEFAULT_BASE+ assert descriptor.base_url_attr is None+ assert descriptor.proxy_attr == "atlascloud_proxy"+ assert config.api_key == "atlascloud-token"+ assert config.base_url == ATLASCLOUD_DEFAULT_BASE+ assert config.proxy == "http://proxy.test:8080"+ assert isinstance(provider, OpenAIChatProvider)++@pytest.mark.asyncioasync def test_xai_provider_config_uses_key_base_and_proxy() -> None:descriptor = PROVIDER_CATALOG["xai"]@@ -1172,6 +1205,7 @@ async def test_create_provider_instantiates_each_builtin():"experiential": OpenAIChatProvider,"cheaperinference": OpenAIChatProvider,"orcarouter": OpenAIChatProvider,+ "atlascloud": OpenAIChatProvider,"xkiro": OpenAIChatProvider,"opencode_go": OpenCodeProvider,"vercel": OpenAIChatProvider,